Re: large negative parameter correlations in regression

Reef Fish wrote:
Anon. wrote:
Reef Fish wrote:
Anon. wrote:
Reef Fish wrote:
Anon. wrote:
Reef Fish wrote:
Stephen Clark wrote:
<snip>
So, no change there.
--
Bob O'Hara
Department of Mathematics and Statistics
Your reply was so non-specific, so speculative, so name-dropping
without knowing what you were talking about that it would require
FREE tuition to explain it to YOU, without benefiting the OP.

Name dropping? Check: I didn't even mention any names!

Name dropping don't always refer to PERSONAL name-dropping.
You were dropping names of "statistical terms and procedures"
of which you had no idea of the relevance or irrelevance of them
to the OP's question. Such as "prediction", "covariates",
"correlation" (don't even know which of the 4 different correlations
you were talking about), etc. etc.

Non-specific? Like this: "Prediction: the mean of the variable is a
long way from zero." That's so specific, it's falsifiable.

I had classfied that as "speculative", very specifically speculative,
and WITHOUT any reason given for the speculation.

Of course it's falsifiable. Why open your mouth? You had NOTHING
tangible or useful to offer other than your unsupported speculation.

That's exactly the kind of comments you always throw in, when
you step into a subject in which you have no useful knowledge to
offer.

Or like
this: "If you mean-centred your variables, then the correlation will go
away." Again, falsifiable.

It's not only falsifiable, it is FALSE.

OK, prove it.

Go on, show that the second statement is wrong, i.e that if the mean of
a covariate in a simple least-squares regression is zero, then the
correlation between the slope and intercept is non-zero.

Then you would NOT have an intercept, you IDIOT!

What? The model would be

y_i = a + b x_i + e_i

a is the intercept. Why does a=0 if sum x_i = 0?

Bob

--
Bob O'Hara
Department of Mathematics and Statistics
P.O. Box 68 (Gustaf Hällströmin katu 2b)
FIN-00014 University of Helsinki
Finland

Telephone: +358-9-191 51479
Mobile: +358 50 599 0540
Fax: +358-9-191 51400
WWW: http://www.RNI.Helsinki.FI/~boh/
Journal of Negative Results - EEB: www.jnr-eeb.org
.

Relevant Pages